Adjustment for a linear session trend changed theta power versus current slope signs in three of four people across 149 hippocampal depth records

We tested whether programmed current and pre-pulse theta power have a stable relation in public implanted depth records. We used 250 records from seven people in OpenNeuro ds007095, version 1.0.0. Four people had current changes across 149 retained records. A linear session adjustment changed individual slope signs in three of these four people. Current and session order were strongly related. The primary common slope after person and montage adjustment was 0.069 log10(µV²) per mA. Its 95% person bootstrap range was −0.140 to 0.128. The planned change from the balanced pooled slope was −0.117, with a range of −0.668 to 0.250. The primary comparison did not establish a decrease. A secondary near-pulse trend-adjusted range was positive. The data do not identify a causal current effect or a stable dose rule. The package preserves 250 source EDF records, sidecars, source identities, new code, all results, three figures, and reproduction instructions. Independent signal and regression checks passed. A second run reproduced all four numeric files exactly. These are chronic hippocampal depth records, labelled SEEG in recording sidecars and DBS in channel sidecars. The cohort overlaps earlier sci work.

responsive neurostimulation · theta power · hippocampal depth recordings · session trend · person-level analysis · current amplitude · brain interfaces